Skip to main content
POST
/
v0
/
models
Create a custom model
curl --request POST \
  --url https://api.tokenfactory.nebius.com/v0/models \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "name": "<string>",
  "base_model": "<string>",
  "source": "<string>",
  "description": "<string>"
}
'
{
  "name": "<string>",
  "base_model": "<string>",
  "source": "<string>",
  "created_at": 123,
  "status": "validating",
  "description": "<string>"
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Query Parameters

​
ai_project_id
string | null

Project ID to list models from.

Body

application/json
​
name
string
required

The new model name.

Example:

"my-custom-model"

​
base_model
string
required

The base model identifier the new model belongs to

Example:

"meta-llama/Meta-Llama-3.1-8B-Instruct-fast"

​
source
string
required

Checkpoint id from Fine-Tunes API, file id from Files API, or HuggingFace repo url

Examples:

"ftjob-1234567890:ftckpt_1234567890"

"file-1234567890"

"huggingface.co/sofrony/Llama-3.1-8B-Instruct-sofrony"

​
description
string | null

Response

Successful Response

​
name
string
required

The new model name.

Example:

"meta-llama/Meta-Llama-3.1-8B-Instruct-LoRa:my-custom-model-udpd"

​
base_model
string
required

The base model identifier the new model belongs to

Example:

"meta-llama/Meta-Llama-3.1-8B-Instruct-fast"

​
source
string
required

Checkpoint id from Fine-Tunes API, file id from Files API, or HuggingFace repo url

Examples:

"ftjob-1234567890:ftckpt_1234567890"

"file-1234567890"

"huggingface.co/sofrony/Llama-3.1-8B-Instruct-sofrony"

​
created_at
integer
required
​
status
enum<string>
required

The status of the model

Available options:
validating,
active,
error,
deleted
Example:

"validating"

​
description
string | null